The autoimmune destruction of insulin producing cells in IDDM is critically controlled by the conserved CD1d-autoreactive NKT cells. NKT cells express an invariant TCR alpha chain combined with a restricted set of TCR Vbeta families (Valpha14- Jalpha18/Vbeta8 in mouse and Valpha24-Jalpha18/Vbeta11 in humans) as well as a panoply of receptors of the NK lineage. Upon in vivo stimulation, they promptly release TH2 cytokines that suppress diabetes. We have developed a set of tools, such as CD1d/alphaGalCer tetramers specifically staining the semi- invariant NKT cell TCR, which allow us for the first time to specifically study NKT cells in vivo in both mice and humans. This application proposes to dissect and compare the mechanisms underlying the NKT cell defects associated with IDDM in mice and humans. The specific aims are Specific Aim number 1: To identify the defective stages and mechanisms of Valpha14 NKT cell development in NOD mice. We will dissect the defect in NKT cell thymic and post-thymic development, and identify the defective cellular and molecular mechanisms underlying the NOD phenotype. In particular, the nature of the defective cell-type will be dissected using a mixed chimera system in vivo, and the role of IL-15, a key cytokine in NKT cell development, will be investigated in vitro and in vivo. Specific Aim number 2: To determine the contribution of CD1d and Valpha14 T cells to NOD mouse diabetes. We will study NOD.CD1d0 mice and their complementation cross with NOD.CD1d-Tg mice, as well as NOD.Jalpha180 mice, in order to test the hypothesis that NKT cells naturally regulate spontaneous diabetes. We will perform transfers of Valpha14 NKT cells in various combinations of mutant donor (e.g. IL-40) and host (e.g. CDld0) mice to dissect the functional basis of the protective effect. Specific Aim number 3: To identify the defective stages of Valpha24 NKT cell development in IDDM patients and their at risk relatives. We will define the human correlates of the mouse NKT cell subsets and dissect their defects in diabetic patients and at risk relatives. We will determine the timing and context of the NKT cell failure with respect to other established immunological markers of the disease process. These combined studies of mouse and human NKT cells have the potential to lead to novel strategies for the prediction and prevention of autoimmune diabetes in humans at risk for this disease.